A Service of

Leibniz-Informationszentrum econstor Wirtschaft Leibniz Information Centre Make Your Publications Visible. zbw for Economics

Misra, Joya; Moller, Stephanie

Working Paper Famililism and welfare regimes: Poverty, employment, and family policies

LIS Working Paper Series, No. 399

Provided in Cooperation with: Luxembourg Income Study (LIS)

Suggested Citation: Misra, Joya; Moller, Stephanie (2005) : Famililism and welfare regimes: Poverty, employment, and family policies, LIS Working Paper Series, No. 399, Luxembourg Income Study (LIS), Luxembourg

This Version is available at: http://hdl.handle.net/10419/95502

Standard-Nutzungsbedingungen: Terms of use:

Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Documents in EconStor may be saved and copied for your Zwecken und zum Privatgebrauch gespeichert und kopiert werden. personal and scholarly purposes.

Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle You are not to copy documents for public or commercial Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich purposes, to exhibit the documents publicly, to make them machen, vertreiben oder anderweitig nutzen. publicly available on the internet, or to distribute or otherwise use the documents in public. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, If the documents have been made available under an Open gelten abweichend von diesen Nutzungsbedingungen die in der dort Content Licence (especially Creative Commons Licences), you genannten Lizenz gewährten Nutzungsrechte. may exercise further usage rights as specified in the indicated licence. www.econstor.eu

Luxembourg Income Study Working Paper Series

Working Paper No. 399

Familialism and Welfare Regimes: Poverty, Employment and Family Policies

Stephanie Moller and Joya Misra

January 2005

Luxembourg Income Study (LIS), asbl

Familialism and Welfare Regimes: Poverty, Employment, and Family Policies

Joya Misra Stephanie Moller University of Massachusetts-Amherst University of North Carolina-Charlotte Social and Demographic Research Institute Department of Sociology Machmer Hall – 240 Hicks Way 9201 University City Boulevard Amherst, MA 01003 Charlotte, NC 28223

(413) 545-5969 (704) 687-3237 Fax (413) 545-0746 Fax (704) 687-3091 [email protected] [email protected]

Joya Misra is an Associate Professor of Sociology and Public Policy at the University of Massachusetts Amherst. She has recently published papers in Social Problems, the Annual Review of Sociology, Socio-Economic Review, and Research in Political Sociology. Her current research includes an analysis of the effects of family policy in explaining the motherhood wage gap in a variety of countries, and a comparative project on immigration, welfare state restructuring, and the international division of care.

Stephanie Moller is Assistant Professor of Sociology at the University of North Carolina at Charlotte. She has recently published "Supporting Poor Single Mothers: Gender and Race in the US Welfare State (Gender and Society, 16, no. 4, 2002) and coauthored "Determinants of Poverty in Advanced Capitalist Democracies" (American Sociological Review, 68, no. 1, 2003). Her current research examines state-level determinants of stratification.

An earlier version of this paper was presented at the 2003 annual meetings of the Social Science Historical Association, and in the 2003-2004 Center for Public Policy and Administration Colloquium Series. This material is based upon work supported by the National Science Foundation under Grant #SES-0095251, the Social and Demographic Research Institute, and the Center for Public Policy and Administration at the University of Massachusetts-Amherst. Thanks to assistance provided by Jessica Cichalski, Karen Mason, and Sabine Merz. We appreciate helpful comments from many scholars, particularly, Eric Einhorn, Nancy Folbre, Barbara Hobson, Ann Orloff, and Dylan Riley.

Familialism and Welfare Regimes: Poverty, Employment, and Family Policies

Abstract

While many nations lay a claim to supporting “family values,” these values may be interpreted in

a variety of ways. How do nations support families, particularly families with children? What

strategies do different nations take, and how do these strategies lead to different outcomes? In

this paper, we show how different combinations of policies that support family caregiving and

those that de-familialize caregiving lead to significantly different outcomes. We show that

nations with stronger levels of both kinds of policies have lower poverty levels than those with weaker levels of these policies, but that strong levels of policies that support family caregiving and weak levels of de-familializing policies have more varied results, with higher levels of poverty, particularly for families headed by single mothers. In addition, this research illustrates significant variation among “continental Conservative” countries, and suggests the importance of a less static approach to welfare state regimes, which also fully recognizes the centrality of gender relations to labor market and welfare state policies.

2 Familialism and Welfare Regimes: Poverty, Employment, and Family Policies

Over the last several decades, advanced industrialized countries have changed in notable ways: welfare states have undergone restructuring, women have entered the labor force in growing numbers, and family forms have experienced serious alterations. A critical array of family policies has played a key role in addressing these changes and has helped balance the needs of families with the operation of both states and markets. For example, effective family policies may address families’ needs by substantially lowering poverty rates for families with children, and in particular for single parent families (Bradshaw et al. 1993; Folbre 1994;

McLanahan and Garfinkel 1995; Rainwater and Smeeding 2003). Family policies may also impact the family-market nexus by shaping whether women primarily pursue employment or remain in the home and provide caretaking for their families when children are young (Folbre

1994; Gornick and Meyers 2003). As O’Connor, Orloff, and Shaver (1999, p. 1) argue,

Debates about the proper role of the state vis-à-vis the market and family and about the

character of state policies have intensified and broadened out to consider a greater range

of policy alternatives. . . Concerns of gender pervade these social policy debates – about

employment opportunities and day care, about how (and even whether) to publicly

support caregiving work and single parent families, about the scope of women’s choices

as to whether and when to be mothers.

In this study, we explore the strategies that nations take to support families, and how these strategies lead to different outcomes, particularly regarding poverty and employment, for both single and married mothers. We explore how the intersection of familialistic policies

(supporting family caretaking) and de-familializing policies (undermining family caretaking) leads to different constellations of policies, and how these constellations reflect and lead to different levels of women’s labor force participation. By relating poverty rates for a variety of families – including married and single, childless, with children, and with young children – we assess the effectiveness of these policies regarding poverty prevention. In addition, we analyze how these policies impact married and single mother families, because single parent families

“raise fundamental issues about the balance between family, state, and individual financial responsibilities; and about the roles of men and women as parents and workers” (Millar 1996, p.

25).

In addition, our analysis reconsiders the welfare regime approach, and how, in particular, the “continental European Conservative” regime, which is said to be particularly familialistic, operates as a policy strategy. Here, we focus upon Belgium, France, Germany, and the

Netherlands. We show that there are important variations among these nations, and that these differences may point to a more dynamic model of welfare state regimes. While the welfare regime concept has led to very powerful and effective analyses of welfare states, this concept assumes that welfare states hold a fixed position over time. Yet with the dramatic changes in gender relations, welfare state policies have changed in substantial ways, leading to substantial variations within regime type. A model that better captures these transformations would be preferable.

This article begins with a discussion of the welfare regime concept and, in particular, critiques focused on familialism and de-familialization. From there, we place the Conservative continental countries within this theoretical context. Next, we examine how nations cluster regarding poverty rates and women’s labor force participation rates for a variety of family types.

We examine some potential explanations for these different clusters, emphasizing the

2 intersection of policies that support and those that destabilize family caregiving. Finally, we

draw out the implications of our findings for studies of welfare regimes.

Welfare Regimes, Familialism, and De-Familialization

For comparativists, welfare state scholars over the last fifteen or so years have focused on explaining differences between “families of nations” (Castles 1993; Castles and Mitchell 1993) and “worlds of welfare capitalism” (Esping-Andersen 1990; Kolberg and Esping-Andersen

1992), showing how nations tend to cluster in certain groups in terms of policy creation and outcomes. The idea behind grouping nations in certain broad categories is that we can see qualitative differences between groups in the origins of social policies and their outcomes, which

can then help us identify the different strategies nations take.

Esping-Andersen’s (1990) very influential welfare state regime typology is often used to

place countries into one of three diverse regime types. In the Liberal regime (e.g., Australia,

Canada, the United Kingdom, the United States), the state avoids policy measures that tamper

with the market. In the Conservative regime (e.g., France, Germany, Italy), the Church-shaped

state uses policy to uphold status differences and preserve the traditional family. In the Social

Democratic regime (e.g., Denmark, Norway, Sweden), the state uses policy to redistribute wealth

and support full employment. Esping-Andersen (1990, p. 22) originally grouped countries into

these clusters based on state-market relations, stratification, and social citizenship rights,

including their level of de-commodification, or how state policies allow citizens to “maintain a

livelihood without reliance on the market.” De-commodification was measured by the

generosity and availability of old age pensions, sickness benefits, and unemployment insurance

payments.

3 Feminist scholars have roundly critiqued this model for focusing on de-commodification, particularly insofar as this model does not account for women’s experience within the welfare state. Bussemaker and van Kersbergen (1994, p. 13) note that the regime approach focuses on class inequality, state provision of welfare, and the de-commodification of wage-labor, underrating gender inequalities and “analytically neglect[ing] the role of the family (and women in particular)” in care. As a result, Esping-Andersen’s regimes do not always explain variations in policies directed at families, rather than those targeting workers, and do not adequately predict women’s labor force participation (Langan and Ostner 1991; Taylor-Gooby 1991; Orloff 1993,

1996; Hobson 1994; Sainsbury 1995). For example, his conceptualization suggests that women’s work historically has been marginalized in Conservative countries, and that welfare policies encourage women to remain in the home to preserve the traditional family structure

(Esping-Andersen 1990; Castles 1993; Gornick et al. 1997). Yet, even within a particular regime type, there may be significant differences in levels of women’s employment.

Lewis and Ostner (1991) provide an alternative typology that divides states among strong, moderate, and weak breadwinner ideologies. According to their scheme, in certain

“strong breadwinner” nations, men are seen primarily as breadwinners and women as caretakers.

In a moderate breadwinner state, women have somewhat more accepted roles as workers outside the home. In the weak breadwinner state, women and men are viewed as equally involved in caretaking and working outside the home. State policies support these different roles, for example through tax policies that reward single breadwinner families (Ostner 1994).

Jane Millar (1996) argues that even the male breadwinner model does not accurately capture gendered experiences, because welfare states may reinforce gender roles differently for married and single mothers (Schultheis 1992; Hobson 1994; McLanahan & Garfinkel 1995;

4 Lewis 1997; Hobson and Takahashi 1997; Ostner 1997; Bussemaker et al. 1997; Kiernan et al.

1998; Bussemaker 1998). Millar (1996) suggests that countries can be categorized in three

ways: those which encourage employment for both single and married mothers; those which

discourage employment for both single and married mothers; and those which discourage

employment for married mothers, and encourage employment for single mothers. The first

grouping of countries encourages employment for both groups by providing services such as

childcare, employment rights such as parental leave and equal pay policies, and benefits. The second group discourages employment for both groups by providing lower levels of support for employment, and weaker benefits. As a result, women tend to work part time, and single mothers receive social benefits since mothers are not expected to work full time. Finally, the third group discourages employment for married mothers by not providing any supportive services, while encouraging employment for single mothers not through positive measures, but through a lack of alternatives such as social benefits.

Ann Orloff (1993) specifies how Esping-Andersen’s model, and power resource theory more generally, should be reconceived to include attention to not only class but also gender

effects. She argues that rather than simply focusing on state-market relations in providing

welfare, scholars must understand that the family is a central actor in the provision of welfare.

With this understanding, we can go on to explore how “[s]tate provisioning that helps to shift the

burden of welfare from the family to the state, or from women to men within the family furthers

women’s gender interests” (Orloff 1993, p. 312). Similarly, rather than focusing on how social

provision mediates or reifies class stratification, we must understand how social provision also

mediates or reifies gender stratification. Social citizenship rights are different for women

workers, who may not be commodified in the same way men are while also being relied upon to

5 provide domestic and caretaking work. By ignoring the tremendous amount of unpaid labor that

many women do within the family, how this work structures women’s lives, and how welfare programs target women through their family roles, Orloff claims that Esping-Andersen’s model is incomplete. While recognizing that there are multiple interpretations of how to address these gendered inequalities, Orloff argues for a welfare state model that attends to women’s access to paid employment and women’s capacity to form and maintain autonomous households.

Esping-Andersen (1999) has responded to these critiques by including in his later regime conceptualization the ideas of de-familialization and familialism. De-familialization refers to

“the degree to which households’ welfare and caring responsibilities are relaxed – either via

welfare provision or market provision,” while familialism refers to a system where “public

policy assumes – indeed insists – that households must carry the principal responsibility for their

members’ welfare” (Esping-Andersen 1999, p. 51). Familialism is strongly influenced by

Catholic social teaching, particularly the principal of subsidiarity, which argues that the state and

community should not undermine the autonomy and responsibility of the family. Esping-

Andersen then measures de-familialization by examining public spending on family services, the

percentage of children under three in childcare, and the percentage of elderly receiving public

home help. He goes on to show how these measures relate to the intensity of family welfare

provision, based on the percentage of the aged living with their children, unemployed youth

living with parents, and women’s weekly unpaid hours. Incorporating this measure of de-

familialization, however, does not drastically change the original regime concept. Indeed, the

Liberal regime countries continue to cluster together, and most Social Democratic countries

remain clustered together, while clearer differences emerge between the countries originally

defined as “Conservative.”1 Here he finds two groups – continental European countries (Austria,

6 Belgium, France, Germany, and the Netherlands) and southern European countries (Italy,

Portugal, Spain) – with southern European countries scoring the lowest on de-familialization.

Sigrid Leitner (2003) extends this concept further with her categorization of types of familialization and de-familialization. Leitner agrees that there are both familialistic policies, which attempt to strengthen the family in its caring role, and de-familializing policies, which work to relieve the family of providing direct care. For example, familialistic policies might include parental or family leaves, cash benefits or tax reductions for caregivers, or social rights attached to care, such as pensions. Such policies strengthen the family’s caring role, which may reinforce women’s role in care, although policies may also be written to encourage men’s caring roles. De-familializing policies might include the public provision of childcare, elder care, or other social services. While de-familializing policies may not directly intervene in gender relations, by relieving the family from care they may weaken male breadwinner models.

According to Leitner, welfare states combine familialistic and de-familializing policies. She develops the typology illustrated in Table One.

[Table One About Here]

Optional familialism is characterized by those countries with both services for caretakers and supportive care policies for workers; explicit familialism by those countries with supportive care policies but no services; de-familialization by those countries with services but no supportive care policies; and implicit familialism by those countries with neither services nor supportive care policies. Leitner (2003) analyzes how different policy arrangements in EU member states suggest different forms of familialism and de-familialization. For example, counter to Esping-

Andersen’s (1999) argument, Leitner argues that Social Democratic nations appear to provide optional familialism rather than de-familialization. Similarly, Liberal countries appear to

7 provide both explicit familialism and de-familialization; and Conservative countries are split between explicit and implicit familialism. Leitner goes on to analyze how these different policies reinforce traditional gender roles, by looking at how these values include assumptions about men’s and women’s gender roles in regard to parental leave.

However, Leitner’s model remains problematic. For example, the United Kingdom and

Ireland are categorized as de-familializing because formal childcare is “widespread,” in that a little more than one-third of children under 3 are in formal child care, while there is no paid parental leave policy. This is because Leitner’s model is based on private or public childcare; the UK and Ireland do not subsidize childcare, but leave it to the market – making it less available for many parents (Gornick and Meyers 2003). However, Leitner does provide another important way to consider familialism.

Conservative Countries and Familialization

Despite the criticisms of Esping-Andersen’s (1990) original model, many welfare state scholars continue to explore how countries within regimes share similar approaches or diverge considerably. While substantial research focuses upon variations in Liberal regime countries

(O’Connor et al. 1999) and Social Democratic regime countries (Erikson et al. 1987; Leira

1992), less research explicitly focuses on variations among Conservative countries (Daly 1999,

2000). Yet, the Conservative regime should be explored more thoroughly, particularly since support for family caregiving may play an important role in shaping welfare policy in these nations (Kaufmann 1988; Lessenich and Ostner 1995; Daly 1999, 2000; Esping-Andersen 1999;

Leitner 2003).

8 According to Esping-Andersen (1990), Conservative countries are distinct from Social

Democratic and Liberal countries in part because of the focus on the traditional family, which limits women’s labor force participation. As he notes:

In most other countries, the modern welfare state was built by Social Democrats or left

Liberals; welfare state consolidation in postwar Europe was dominated by rightist or

centre-right coalitions – Christian Democracy in particular. Even where Christian

Democratic parties were marginal, such as France or Spain, Catholic social doctrines still

exert a visible influence on social policy (1996, p. 66).

Yet although Catholic social doctrine meant to support traditional family values shapes these

Conservative countries, the resulting policies do not necessarily lead to higher fertility rates.

Esping-Andersen argues that paradoxically “familialistic policy appears counter-productive to family formation” (1999, p. 67), noting that the de-familialized Social Democratic nations have higher fertility rates than the highly familialized southern European nations. Finally, Esping-

Andersen (1999) argues that among the original Conservative regime types, continental

European countries, such as Belgium, France, Germany, and the Netherlands, differ from the southern European countries, such as Italy, Portugal, and Spain, in their levels of familialism and de-familialization. The Mediterranean countries have extremely high levels of familialism, which has consequences both in terms of lower levels of services and benefits and in terms of women’s employment. The continental Conservative countries, however, provide a closer match to the theoretical expectations of the Conservative regime.

We examine whether there are further divergences within the continental European countries, and what such divergences might indicate. We focus on the cases of Belgium, France,

9 Germany, and the Netherlands. In terms of our theoretical framing, the classification of these

cases is explicated in Table Two.

[Table Two About Here]

Belgium, France, Germany, and the Netherlands have been identified as Conservative nations

(although the Netherlands has occasionally been typed as Social Democratic) (Esping-Andersen

1990, 1996, 1999; Hobson 1994; Knijn 1998; Van Kersbergen and Becker 1988). More recently,

these nations have been further specified by Esping-Andersen as “continental European”

Conservative nations, which are different in approach from the even more familialistic southern

European nations. His conceptualization suggests that these four nations are fairly similar.2

According to Lewis and Ostner’s (1991) typology, Germany and the Netherlands can be

typed as “strong breadwinner” states, where men are seen primarily as breadwinners and women

are seen as caretakers. However, France and Belgium are “moderate breadwinner” states, where

women have somewhat more accepted roles as breadwinners (also see Pfau-Effinger 1999). The

state uses social policies to shape these differences. For Lewis and Ostner, Esping-Andersen’s regime types clearly do not adequately capture the difference in women’s workforce participation among these nations.

Millar’s (1996) critique argues that rather than simply looking at male breadwinning, we must also examine how policies influence both married and single mothers to work or caretake.

She suggests that policies do not simply influence men’s versus women’s roles as breadwinners or caretakers, but may reinforce different roles for women based on whether they are married or single. Millar categorizes countries in three ways: those with high employment rates for both single and married mothers (e.g., France, Belgium); those with low employment rates for both single and married mothers (e.g., the Netherlands); and those with low employment rates for

10 married mothers and high employment rates for single mothers (e.g., Germany). However,

Belgium appears to actually have somewhat lower employment rates for single mothers, relative to other nations, suggesting that there may potentially be a fourth category, with low employment rates for single mothers and high employment rates for married mothers. Here, as with the breadwinner typology, it appears that there are important differences among the continental Conservative countries.

Building on Ann Orloff’s (1993) concept of “de-familialization” and Esping-Andersen’s revision of his framework to analyze familialism and de-familialization, Leitner (2003) suggests that famialism and de-famialization may combine to lead to a number of different outcomes.

Leitner finds somewhat different findings by country, depending on whether she focuses on elder care or childcare. Regarding childcare, Belgium and France could be typed as promoting

“optional familialism,” while Germany and the Netherlands are typed as “explicit familialism.”3

Leitner again suggests that there is significant variation among the Conservative continental countries.

There do seem to be some significant differences among Belgium, France, Germany, and the Netherlands. Belgium and France are classified together in each of the theories. Germany and the Netherlands are also similarly classified, but they are more distinctive, particularly in terms of Millar’s classification. Understanding these differences may help reconceptualize how we group countries, and what the underlying assumptions are behind the strategies different countries take.

Familial Outcomes

While the theoretical models provide a number of useful approaches to thinking about how welfare policies shape famialialism and de-familialization, we should also incorporate

11 measures of the outcomes regarding women’s dependence. We highlight the differences in outcomes among Social Democratic, Liberal and Christian Democratic countries, including

Belgium, France, Germany, the Netherlands, Finland, Norway, Sweden, Canada, the United

Kingdom, and the United States. We focus first on poverty rates and then on women’s labor force participation rates. Finally, we discuss the qualitative differences between policies across countries, particularly the Christian Democratic countries, to help explain variations in outcomes.

We explore these outcomes by examining poverty rates for women with and without children. Poverty rates for women may be lower both where women have greater labor market opportunities and where women receive provisioning through the state to support their caregiving roles (Orloff 1993). We use the Luxembourg Income Study (LIS) database to develop a number of measures of poverty rates across families with and without children. We rely on LIS data because this database provides the best cross-national data for comparing income across OECD countries (OECD 1995). LIS gathers data from national surveys and harmonizes the data to ensure comparability. The data are organized into waves. We utilize data from Waves IV and V to compare poverty rates across the Conservative Democratic countries.

Wave IV includes data for the mid-1990s, and Wave V represents the early 2000s (see tables for exact years). Wave V data are not yet available for Belgium and France.

Like most comparative researchers, we measure poverty rates relatively to capture the extent that families fall below 50 percent of their countries’ median income (Brady 2003; Casper et al. 1994; Korpi and Palme 1998; Moller et al. 2003). We constrain our sample to working age adults to limit the number of students and pensioners in the sample. Pensioners are highly dependent on transfers. Thus, including them in our analyses overstates the extent of

12 redistribution and undermines our ability to examine the link between state policies and markets

in terms of families (Huber et al. 2003; Moller et al. 2003).

We examine three measures of poverty. Pre-tax and transfer poverty rates represent

market-based income. They are calculated as the percentage of households with working age

adults, 25-59, with market incomes below 50 percent of median income. Market income

includes total household income from wages and salaries, property income, private pension

income, and self-employment income.4 Post-tax and transfer poverty rates represent total

income including governmental transfers. We calculate post-tax and transfer poverty rates as the

percentage of households headed by working age adults, 25-59, with disposable incomes below

50 percent of median income. Disposable income includes market income, governmental transfers, and taxes.5

To examine how state policies affect the poverty rates of families, we further

disaggregate our sample by family type and the presence of children. This disaggregation is

necessary because understanding single mothers’ experiences with poverty may be crucial. As

Orloff (1993) argues, a central component to welfare state provision may be women’s capacity to

form and maintain autonomous households. If single mothers face significantly higher levels of

poverty, this capacity is clearly undermined.6

Following Karen Christopher (2002), we calculate poverty rates separately for

households headed by married and single mothers. We define married couple families broadly to

include cohabitating and married couples. All unmarried, separated, or divorced women who do

not live with an adult of the opposite sex are categorized as “single.” We also distinguish

between women without children under 18, mothers with children under 18, and those with

children under 6, although the data for families with children under 6 is based on fewer cases and

13 may not be as reliable for single mothers.7 We present the number of cases in the calculations of

pre-tax and transfer poverty rates in Appendix A. We begin by examining pre-tax and transfer

poverty levels for married and single women in households without children, with children under

18, and with children under 6, before any tax and transfer programs are put into place. Table

Three summarizes these data.8

[Table Three About Here]

First, it should be evident that married women face significantly less poverty than single women. Overall, single women appear three to four times more likely than married women to fall into poverty. This is dramatically true even for women without children. For example, during the most recent period, almost 22% of single women without children are poor on average, compared to 7% of married women without children. Women are not necessarily able to stave off poverty in any of these nations by relying on their wages through the labor market.

Marriage still appears to be the solution to poverty for many women.

Poverty rates are also substantially higher for women with children, particularly when these children are young (under 6). For example, during the most recent wave of data, 7% of married women without children, on average, faced poverty, compared to 11% of married women with children, and 14% of married women with young children. Similarly, on average,

22% of single women faced poverty, compared to 52% of single women with children, and 69% of single women with young children. The presence of children, particularly young children, plays a key role in whether women experience poverty (Folbre 2001). Looking at changes over time, average poverty rates appear to be somewhat lower in many of these nations in the more

recent wave of data, perhaps due to the economic recovery after the recession of the early to mid-

1990s.

14 We can also examine the rankings of countries to discover interesting patterns. Countries are ranked by poverty rate from lowest to highest; difference less than .3 share a rank. Thus, for married women without children in the mid-1990s, Norway had the lowest market-based poverty rates, followed by Sweden. The Conservative countries, in contrast, have the highest poverty rates for married mothers without children. These rates remain high circa 2000; however, there is little variation in childless married women’s poverty rates during this period. Pre-tax and transfer poverty rates for married women with children (and with young children) are the lowest in the Conservative regime, and the worst in the Liberal regime during the 1990s. By the 2000s, the distinction between the Conservative and Social Democratic regimes is minimized by overlapping poverty rates.

The market-based poverty rates for single women without children follow less distinguishable patterns. Among single women with children, Belgium and France have particularly low pre-tax and transfer poverty rates and Germany and the Netherlands have remarkably high rates. This pattern follows Lewis and Ostner’s insights that Belgium and

France are moderate breadwinner countries, while Germany and the Netherlands have a strong breadwinner ideology. It also corresponds with Leitner’s observation that Germany and the

Netherlands offer limited services to permit work among single mothers. Interestingly, this hypothesis does not hold up during the 2000 period because single women households with children in Germany and the Netherlands have lower poverty rates than those households in the

Social Democratic and Liberal nations (excluding Finland).

While this data suggests some interesting differences between countries, it does not capture how tax and welfare programs attempt to ameliorate poverty for these groups. Since we are most interested in how welfare states, or the complex of welfare policies, shape the

15 experiences of families, it is crucial to examine post-tax and transfer poverty rates and reductions

in poverty due to the tax and transfer systems. How do these countries use taxation and welfare

policies to bring families out of poverty? Are families with children less likely to fall into poverty, once we account for transfers? Table Four summarizes post-tax and transfer poverty rates across these countries.

[Table Four About Here]

In most countries, poverty rates drop significantly when taking the effect of welfare and tax policies into account. After tax and welfare policies have done their work, poverty rates drop from 7%, on average, for married childless women to 3.3 %; from 22%, on average, for single childless women to 12%. Perhaps even more impressive, poverty rates for women with young children drop from 14% and 69% for married and single women, respectively, to 7% and 32% for married and single women respectively. While poverty rates for single women, particularly those with children, remain disturbingly high, they are much lower than pre-tax and transfer

(although this drop is less remarkable for the United States).

As in Table Three, these data suggest that marriage remains a key route out of poverty for many women. Married women have much lower levels of poverty both pre- and post-tax and transfer. Perhaps because women’s wages are lower in the labor market, marriage appears to strongly lower poverty rates for women in every nation. Unpartnered women still run a higher risk of living in poverty, whether they have children or not. Indeed, single childless women both pre- and post-tax and transfer are more likely to live in poverty, on average, than married women with young children. Women appear constrained in their ability to form autonomous households without slipping into poverty (Orloff 1993).

16 In addition, having children, particularly young children, continues to increase the chance

of poverty significantly. For example, 3.3% of married women without children live in poverty, on average, compared to 5.2% of married women with children, and 6.7% of married women with young children. Differences in poverty rates based on having children remain, but they are clearly ameliorated somewhat by tax and transfer. Because of the effect of the presence of children and being unmarried, single mothers with children remain vulnerable – on average, 23%

of single women with children and 32% of single women with young children live in poverty.

Yet in addition to understanding the overall trends among groups of women across these

nations, we are most interested in determining whether these nations seem to cluster together in

groups in terms of poverty rates for married and single women. These clusters may provide

some evidence for different kinds of strategies. We see in Table Four that countries in the Social

Democratic regime generally have the lowest overall poverty rates, and those in the Liberal

regime generally have the highest. However, there are differences across regimes. To ease

interpretation of these statistics, we categorize countries as rising above the mean poverty rate

(high) and falling below the mean poverty rate (low) during the earlier period (when we have

more countries available). Here we compare poverty rates for mothers with children under 18,

since the data on poverty rates for mothers with young children are based on substantially fewer

cases.

[Table Five About Here]

Table Five summarizes these post-tax and transfer poverty rates. For some countries,

poverty rates appear somewhat consistent with expectations based on Esping-Andersen’s regime

groupings. Here the Social Democratic countries (Finland, Norway, and Sweden) cluster

together, with lower poverty rates for both single and married mothers. Also, the Liberal

17 countries (Canada, the United Kingdom, and the United States) cluster together, with higher

poverty rates for both single and married mothers. These groupings clearly fall where expected

by Esping-Andersen. Social Democratic countries do appear to use tax and transfer policies to

redistribute income in ways that reduce poverty for both married and single mother families.

Liberal countries do appear to be less generous in their attempts to reduce poverty, relying instead on the market.

Yet as various critics suggest, the continental Conservative countries, such as Belgium,

France, the Netherlands, and Germany, do not group together effectively. France can be clustered with the Social Democratic countries, Germany has higher poverty rates for single mothers and lower for married, Belgium has lower poverty rates for single mothers but higher rates for married mothers, and the Netherlands can be clustered with Liberal countries. This

table provides strong support for Millar’s arguments about the differential effects of welfare and

tax policies on married and single mothers.

Historically, France has taken an active labor market policy approach to women’s

employment that may be more in consonance with the strategies taken by Social Democratic

countries. Germany, on the other hand, has attempted to support traditional marriage, with the

outcome of higher poverty levels of single mothers. Belgium has, like France, taken an active

labor market approach, but perhaps particularly effectively for single mothers. Finally, the

Netherlands, particularly since its recent round of (somewhat ambivalently implemented) welfare

reform, has provided less support for families with children than it once did (Knijn and van Wel

2001). Indeed, as Table Five suggests, the trend between 1994 and 1999 is unmistakable. Dutch

programmatic changes appear to be leading to significantly higher poverty rates, except for

single childless women.9

18

Employment, Family Policies, and Support for Family Caregiving

What might explain the differences we observe, particularly among the continental

Conservative countries? Clearly, these nations cluster in different ways based on poverty rates.

While many of the findings we observe can be explained by the differences in welfare regimes –

for example the redistributive Social Democratic countries versus the market-oriented Liberal

countries – regime differences provide a less compelling explanation for the continental

Conservative nations. These countries vary substantially in their poverty rates for married and single women and mothers.

One explanation may be in differences in women’s employment rates. As the “male breadwinner” typology suggests, women’s labor market engagement varies by country.

Particularly if the tax and transfer system is connected to employment, in order to reward workers, we may find that countries with low levels of women’s employment have significantly higher levels of poverty for women and their families. To explore these possibilities, we highlight differences in women’s labor force participation rates by country in the early 1990s in

Table Six. These data, developed by Saraceno (1997), are unavailable for Denmark, Finland, and Canada. However, they are available for each of the Conservative states, permitting us to determine if these states truly form a cluster, as Esping-Andersen would predict. Again, the data are disaggregated by family type, comparing single mothers to married and cohabitating mothers.

It is also disaggregated by hours of employment, comparing full-time to part-time.

[Table Six About Here]

As Ostner and Lewis’s male breadwinner typology suggests, there are different levels of women’s employment across these countries, particularly when we attend to part-time

19 employment as a share of total employment. In most countries, single mothers are more likely to work full time than married mothers; married mothers are more likely to work part time.

However, in a number of countries, single mothers are less likely to be employed than married mothers, reflecting the constraints single mothers may face balancing employment and caretaking. Full-time employment for married mothers varies from 13% in the Netherlands to

62% in Finland; similarly, for married mothers, full-time employment as a share of total employment varies from 25% in the Netherlands to 89% in Finland. Full-time employment for single mothers varies from 16% in the Netherlands to 67% in France; similarly, for single mothers, full-time employment as a share of total employment varies from 40% in the

Netherlands to 94% in Finland.

Table Seven summarizes overall employment trends. Indeed, Finland, France, and

Sweden, three of the countries that tend to have lower levels of poverty for families with children, also have high levels of employment for mothers, including single mothers. As Table

Six suggests, women in Finland and France are more likely to work full time, while many

Swedish women work part time. Germany, the Netherlands, and the UK, with their higher levels of poverty for families with children, also have lower levels of employment for mothers, and higher levels of part-time employment. Belgium, which consistently showed somewhat lower poverty rates for single parent families relative to other countries, also shows higher levels of employment for single mothers, and lower levels of part-time employment.

[Table Seven About Here]

It is interesting to note that these trends support Millar’s (1996) arguments about France as promoting work for both married and single parents, while the Netherlands promotes caretaking for both married and single parents. However, Millar’s model does not fully fit the

20 German case, where married and single parents both seem to have fairly low levels of

employment, although single mothers are more likely to be employed full time. Belgium, as

noted above, also seems to belong to a fourth category, where married mothers are somewhat

more likely than single mothers to be employed at higher levels.

However, employment clearly does not explain all of the differences we see in poverty

rates. While the United States has unusually high poverty rates for both married and single

mother families, it also has unusually high employment rates, and full-time employment rates,

for both groups. Employment may be part of the story, but it clearly is not all of it.

Policies that Support and Limit Family Caregiving

To highlight the extent to which policies vary in tandem with poverty rates and women’s

labor force participation rates, we explore two different types of family policies: parental leave

policies and childcare policies. In keeping with Orloff’s suggestions, and Esping-Andersen’s revised model, we expect that parental leave policies might promote family caretaking, while childcare policies may promote de-familialization. Following Leitner, we are also interested in how famialialism and de-familialization intersect in these nations.

Janet Gornick and Marcia Meyers (2003) have developed a number of indices, which tap into the strength of various family policies, including parental leave and childcare for children at

different ages. Table Eight summarizes their findings, regarding the strength of family policies in

these nations. Once again, the Social Democratic countries fall where we would expect and in

keeping with the poverty rates for families with children; family policies support families with

children through paid parental leaves and subsidized childcare. Similarly, the Liberal countries

21 generally fall where we would expect, providing very limited support for childcare or parental

leave. What about the continental Conservative countries?

[Table Eight About Here]

All four countries provide quite supportive parental leave policies. These countries

provide between 12 and 16 weeks of paid leave for mothers around the birth or adoption of a

child (Gornick and Meyers 2003). In Belgium, the leave is paid for 15 weeks, with 82% of

wages paid for the first 4 weeks (no ceiling), and 75% of wages paid for the following 11 weeks, with a ceiling. In addition, each parent is entitled to 3 months of full-time leave or 6 months of

half-time leave at a flat-rate benefit level, until their child reaches the age of 4. In the

Netherlands, maternity leave is 16 weeks at full pay, with a ceiling. Each parent is entitled to 3

months of full-time unpaid leave or 6 months of half-time unpaid leave, until their child reaches

the age of 8. In Germany, maternity leave is 100% of wages for 14 weeks; parents may also

share up to 3 years of leave, with 2 years of paid leave, for a means-tested flat rate benefit for

which most families qualify. Paid leave is valid until the child is 2; the third year of unpaid leave

must be used before a child turns 8. Finally, France provides 100% of wages for 16 weeks of

maternity leave with a ceiling; this leave is extended to 26 weeks for third and subsequent

children. Parents may share up to 3 years of a flat-rate parental leave for 2 or more children

through the youngest child’s third birthday (Gornick and Meyers 2003, Table 5.1).

All four countries also provide a paid sick child benefit. In Belgium, the benefit is 100%

of wages for up to 10 days a year. In the Netherlands, the benefit is either minimum wage or

70% of the workers’ wage (whichever is higher) for up to 10 days a year. In Germany, the

benefit is 100% for up to 10 days a year per child for married parents (maximum 25 days/year),

100% of 20 days a year per child for single parents (maximum 50 days/year). Finally, in France,

22 sick child benefits are more limited – a 100% benefit for up to 3 days/year, or 5 days/year for

families with children under 1, or 3 children (Gornick and Meyers 2003, Table 5.2).

These parental leave policies, as noted by Leitner (2003), are familialistic in that they

encourage families to provide care for family members. In addition, the policies may support

gendered familialism, insofar that women are more likely to take off time from work to provide

care. While sick child benefits and parental leave are nominally gender-neutral, maternity leave is not. However, Belgium offers a 3-4 day nontransferable paternity benefit at 100% wages; the

Netherlands offers a 2-day nontransferable paternity benefit at 100% wages; and France offers a

2-week paternity benefit at 100% wages, up to a certain ceiling. Germany does not offer

paternity benefits (Gornick and Meyers 2003, Table 5.3). While paternity benefits in no way

come close to maternity benefits, there seems to be a minor attempt to encourage fathers’ roles in

caregiving in Belgium, France, and the Netherlands.

Childcare policy differentiates these nations more strongly. None of these four countries

provide entitlements for public supported early childhood education for children under 2.

However, France provides entitlements from 2 ½, Germany and Belgium from 3, and the

Netherlands from 4. In Belgium, approximately 15% of children under 1, 42% of children under

3, and 99% of children between 3 and 6 are in publicly financed care. In France, approximately

9% of children under 3 are in crèche and 11% are in école, while 99% of children between 3 and

6 are in publicly financed care. In the Netherlands, 17% of children 3 and under 3, and 99% of

4- and 5-year-olds are in publicly financed childcare. In Germany, 5% of children between 1 and

2, and 77% of children 3 to 5 are in publicly financed childcare, although 80% of these children

are in childcare part time (Gornick and Meyers 2003, Table 7.2).

23 These childcare polices are more de-familializing in approach, since they provide care

outside the home. While the policies are not explicitly directed at women, the effects are quite powerful. With high-quality subsidized childcare available, mothers are much more likely to join the labor force, or make a decision to work full time. Childcare policies and parental leave

surely work in concert; parents turn to childcare after using parental leave. However, in France

and particularly Belgium, the policies encourage parents to enter children into publicly supported

childcare earlier.

Constellations of Family Policies and Their Outcomes

What do these differences in family policies suggest about how countries cluster? Table

Nine summarizes these findings in terms of Leitner’s familialistic and de-familializing

framework. While this model may simplify some of the complexity in differences between, for

example, French and Belgian or Dutch and German policies, it does place the range of countries

in recognizable clusters, and in clusters that connect at least in part with the patterns of post-tax

and transfer poverty rates.

[Table Nine About Here]

In the policies that matter most to maternal employment, France and Belgium share more

in common with the Social Democratic countries; as a result, poverty rates tend to be somewhat

lower. There is strong pressure on women in these countries to work. However, policies in these

nations do not simply de-familialize caring. They provide support for some caring inside the

home as well as services for care outside the home, as families may choose to provide care,

particularly for infants. On the other hand, Germany and the Netherlands have policies that are

24 more clearly familializing, providing support for caring inside the home, without the same

degree of commitment to care outside the home.

Of course, we must be cautious about how assertive we are in arguing that nations form

distinct clusters. While German and Dutch childcare services are not as well developed as

services in France and Belgium, they are clearly better than in many southern European nations, or, for example, the United States. Particularly in the Netherlands, where 99% of 4- and 5-year-

olds are in publicly supported care, it is more useful to think of these nations as falling along a

continuum, rather than in clear dichotomous groupings. In addition, we focus on services and

support for care of children; incorporating measures for care of the elderly might significantly

change these groupings (Leitner 2003).

Despite these provisos, we believe that it is imperative to try to understand how policies

that encourage family caregiving and those that relieve family caregiving responsibilities

combine to result in different patterns of support for families, particularly insofar that these

patterns are linked to differences in poverty rates for families with children. As Table Four

suggests, post-tax and transfer poverty rates remain remarkably high in many nations,

particularly for single mothers with children. However, these poverty rates are much higher in

countries that rely primarily on policies supporting family caregiving. Figure One describes

these trends.

[Figure One About Here]

Clearly, the optional familialism model is most effective at lowering poverty rates among

families with children. Countries with models of explicit familialism also show lower levels of

poverty for married mothers with children, compared to the implicit familialism model, although

these levels are higher than for the optional familialism model. However, neither explicit nor

25 implicit familialism models appear to address poverty effectively for single mothers with

children. As Hobson noted in her analysis of single mothers in a variety of countries, in nations

such as Germany, “Solo mothers have no policy domain as either full-time carers or as paid workers. Within the gender logic of the German policy constellation that is organized around

marital status or paid work, solo mothers are a residual category” (1994, p. 182). In keeping

with earlier arguments about the ability of women to form autonomous households, our findings

suggest that family policies shape the experiences of single and married mothers in different

ways, and that de-familializing policies may be particularly important in allowing all women

greater autonomy.

Conclusion

Our research suggests that it is critical to examine family policies in terms of the

dimensions of familialism and de-familialization and to understand the distinct policy

constellations created by combinations of familialistic and de-familializing policies. We show

that these constellations lead to very different outcomes, with regard to poverty rates, for families

with children and, particularly, for families with children headed by single mothers.

While our research supports Esping-Andersen’s basic regime groupings for Social

Democratic and Liberal regimes, it also suggests that even his revised (1999) understanding of

the Conservative regime does not recognize the important variation among family policies within

these nations. Our findings are in keeping with the suggestions of many of Esping-Andersen’s

critics, particularly in our emphasis that maternal employment may be supported in different

ways, and our findings that policies address the needs of single and married mothers in different

ways.

26 While it is certainly important to recognize that southern European Conservative

countries, such as Spain, Portugal, and Italy, provide significantly less explicit family support,

there is also marked variation among the continental Conservative countries, such as Belgium,

France, the Netherlands, and Germany. While Esping-Andersen (1999) notes that familialism is less dominant in France and Belgium, he does not reconceptualize his “regimes” to reflect this difference, in part because he views it as a minor issue relative to the overall forces driving welfare regime differences.

Historically, these Conservative nations may appear to group together. Social

Catholicism, and its doctrine of subsidiarity, clearly has shaped some of these nations’ policies

(though less so in France, with its history of republican anti-clericalism) (van Kersbergen 1994,

1995; Esping-Andersen 1999). As Ilona Ostner (1994) has argued, social Catholic values, while emphasizing the interdependence of individuals, families, community, and the state, support maintaining the autonomy of these units. Within social Catholicism, the state must not undermine the strength, responsibility, and autonomy of the family. Yet, as suggested by

Anttonen and Sipila (1996), France and Belgium are quite different from the other Conservative regime countries regarding how the state intervenes via family policy. While Esping-Andersen

(1999) suggests that this difference is not significant enough to warrant a rethinking of the typology, we argue that such a reconceptualization remains important.

Esping-Andersen (1999) maintains that the main features of the Conservative welfare regime, including labor market regulation and welfare provisioning, center around the family and protecting the role of the male breadwinner. Yet, our analysis suggests that France and Belgium have effectively replaced the male-breadwinner model with a model that supports and expects women’s employment. Higher levels of de-familialization for Belgium and France are not

27 simply a hiccup in an otherwise smooth regime categorization, but a symptom of larger, more

sustained differences between nations. As Bussemaker and van Kersbergen (1999) note, since

1980, Conservative regime countries have faced changing demographic, social, and economic

situations, which have led to a variety of policy changes. High levels of incorporation of women

into the labor market lead to very different models of labor market regulation and welfare provisioning. Recognizing this difference leads to recognizing the importance of taking a more dynamic, and less static, approach to welfare state regimes. While earlier, France and Belgium more effectively fit with the Netherlands and Germany, since the 1980s the divergences among these countries have become more marked.

Similarly, while in earlier periods Dutch and German policy differed significantly, recent policy changes have led to some degree of convergence between these nations. As Hobson

(1994) has argued, in Germany the male-breadwinner model was instituted in a way that covered

women only through their claims as wives. Single mothers received low levels of benefits and

were pushed into the labor market. In the Dutch model, on the other hand, Hobson argues that a

“mother-carer-citizen” model gave single mothers a social wage to support their caretaking. Yet,

since the 1996 Dutch welfare reform, single mothers are no longer supported in the same way

(Bussemaker and van Kersbergen 1999; Knijn 1994). As a result, poverty rates have increased

dramatically for Dutch single mothers, and the Dutch and Germany policy contexts appear more

comparable.

Taking a welfare regime perspective can be extremely powerful. While we do not call for

disbanding the study of welfare state regimes, we do call for a more flexible, and more dynamic,

approach to welfare state regimes, one that recognizes the unmistakable centrality of gender to

the form of welfare state regimes. While welfare states shape demographic changes, such as

28 levels of women’s labor market participation, demographic changes also shape welfare states

(Huber and Stephens 2000, 2001). The origins of welfare states, whether primarily shaped by

Liberal, Social Democratic, or Christian Democratic parties, continue to affect the logic of welfare state provisioning, but restructuring can occur in ways that blur the boundaries between regimes. In particular, we argue that continuing to focus attention primarily on differences between regimes misses important changes occurring within regimes. As Bussemaker and van

Kersbergen (1999) argue, although historical legacies and institutional arrangements certainly play an important role in how countries deal with gender issues, a variety of other political factors (including women’s movements, perceptions of gender equality, EU directives, fiscal crises, changes in governments, etc.) shape policy responses in important ways.

By placing women’s experiences at the center of our analyses, and examining how welfare state regimes support women’s positions as workers and/or caretakers, differentiated by marital status, we develop a better understanding of how policy constellations reflect conceptions of the state, market, and family. We must be willing to reconceptualize welfare state regimes to better capture welfare states’ dynamic approach to balancing state, market, and family.

29 Endnotes

1. However, Esping-Andersen does suggest that the outcomes in terms of the burden of care is not so different among the Liberal and continental European countries.

2. Esping-Andersen (1999) does note that Anttonen and Sipila (1996) have shown that for caring and services Belgium and France “break ranks” with other Conservative nations. Yet he suggests that this is only a minor aberration, and does not have a notable impact on these countries’ placement within the Conservative welfare regime.

3. However, when she focuses on elder care, Leitner instead types Belgium, France, and

Germany as “explicit familialism” and the Netherlands as “implicit familialism.”

4. We adjust income based on household size to account for resource sharing. We utilize a common equivalence scale: the square root of the number of persons in the household (see

OECD 1995 for an overview of equivalence scales).

5. Since some families rely solely on the state for income, we include households without market income in the creation of pre-tax and transfer poverty. However, we exclude households without disposable income in the creation of post-tax and transfer poverty rates.

6. As Hobson argues, “Solo motherhood is the reflector or rearview mirror for the dynamics of power and dependency – the more difficult and stigmatized solo motherhood is in a society, the greater the barriers against opting out of a bad marriage. From this standpoint, the kinds of state support solo mothers receive can be employed as a barometer of the strength or weakness of social rights of women with families” (1994, p. 176).

7. One problem in dividing women into these groups is that they may not be comparable. For example, when we compared the average age of women in these different groups, we found that

30 “childless” women were somewhat older than women with children; these “childless” women may simply be mothers whose children have grown up.

8. Some of these numbers must be viewed with caution. While we have taken out any

calculations based on fewer than 50 cases, almost all of the data examining poverty rates for

single mothers is based on small numbers of cases (on average, fewer than 1,000 cases for single

mothers with children under 18; on average, fewer than 300 cases for single mothers with

children under 6).

9. Bussemaker and van Kersbergen (1999) also note that despite recent increases, through the

early 1990s poverty among Dutch single mother families was relatively low as a result of

generous social transfer.

31 References

Anttonen, A., and J. Sipila. 1996. “European Social Care Services: Is It Possible to Identify

Models?” Journal of European Social Policy 6: 87-100.

Bradshaw, Jonathan, John Ditsch, Hilary Holmes, and Peter Whiteford. 1993. “A Comparative

Study of Child Support in Fifteen Countries.” Journal of European Social Policy 3: 255-

71.

Brady, David. 2003. “The Politics of Poverty: Left Political Institutions, the Welfare State, and

Poverty.” Social Forces 82: 557-88.

Bussemaker, Jet. 1998. “Rationales of Care in Contemporary Welfare States: The Case of

Childcare in the Netherlands.” Social Politics 5: 70-96.

Bussemaker, Jet, Annemieke van Drenth, Trudie Knijn, and Janneke Plantenga. 1997. “Lone

Mothers in the Netherlands.” Pp. 96-120 in Lone Mothers in European Welfare

Regimes: Shifting Policy Logics, ed. J. Lewis. London: Jessica Kingsley Publishers.

Bussemaker, Jet, and Kees van Kersbergen. 1994. “Gender and Welfare States: Some

Theoretical Reflections.” Pp. 8-25 in Gendering Welfare States, D. Sainsbury. London:

Sage.

_____. 1999. “Contemporary Social-Capitalist Welfare States and Gender Inequality.” Pp. 15-46

in Gender and Welfare State Regimes, ed. D. Sainsbury. Oxford: Oxford University

Press.

Casper, Lynne M., Sara S. McLanahan, and Irwin Garfinkel. 1994. “The Gender-Poverty Gap:

What We Can Learn from Other Countries.” American Sociological Review 59: 594-605.

Castles, Francis G., ed. 1993. Families of Nations: Patterns of Public Policy in Western

Democracies. Brookfield: Ashgate Publishing Company.

32 Castles, Francis G., and Deborah Mitchell. 1993. “Worlds of Welfare and Families of Nations.”

Pp. 93-128 in Families of Nations: Patterns of Public Policy in Western Democracies,

ed. F. Castles. Brookfield: Ashgate Publishing Company.

Christopher, Karen. 2002. “Caregiving, Welfare States, and Mothers’ Poverty.” Pp. 113-128 in

Child Care and Inequality: Rethinking Carework for Children and Youth, ed. F. M.

Cancian et al. New York: Routledge.

Daly, Mary. 1999. “The Functioning Family: Catholicism and Social Policy in Germany and

Ireland.” Comparative Social Research 18: 105-33.

_____. 2000. The Gender Division of Labor:The Impact of the British and German Welfare

States. New York: Cambridge University Press.

Erikson, Robert, Erik Jorgen Hansen, Stein Ringen, and Hannu Uusitalo. 1987. The

Scandinavian Model: Welfare States and Welfare Research. London: M. E. Sharpe.

Esping-Andersen, Gosta. 1990. The Three Worlds of Welfare Capitalism. Cambridge: Polity.

_____. 1996. “Welfare States without Work: The Impasse of Labour Shedding and Familialism

in Continental European Social Policy.” Pp. 66-87 in Welfare States in Transition, G.

Esping-Anderson. London: Sage.

_____. 1999. Social Foundations of Postindustrial Economies. New York: Oxford.

Folbre, Nancy. 1994. Who Pays for the Kids? New York: Routledge.

____. 2001. The Invisible Heart. New York: New Press.

Gornick, Janet, and Marcia Meyers. 2003. Families that Work: Policies for Reconciling

Parenthood and Employment. New York: Russell Sage.

33 Gornick, Janet, Marcia Meyers, and Katherine Ross. 1997. “Supporting the Employment of

Mothers: Policy Variation Across Fourteen Countries.” Journal of European Social Policy

7: 45-70.

Hobson, Barbara. 1994. “Solo Mothers, Social Policy Regimes, and the Logics of Gender.” Pp.

170-188 in Gendering Welfare States, ed. D. Sainsbury. London: Sage.

Hobson, Barbara, and Mieko Takahashi. 1997. “The Parent-Worker Model: Lone Mothers in

Sweden.” Pp. 121-139 in Lone Mothers in European Welfare Regimes:Shifting Policy

Logics, ed. J. Lewis. London: Jessica Kingsley Publishers.

Huber, Evelyne, and John Stephens. 2000. “Partisan Governance, Women’s Employment, and

the Social Democratic State.” American Sociological Review 65: 323-42.

_____. 2001. Development and Crisis of the Welfare State: Parties and Policies in Global

Markets. Chicago: University of Chicago Press.

Huber, Evelyne, John D. Stephens, David Bradley, Stephanie Moller, and François Nielsen.

2003. “The Economic Position of Women and the Welfare State.” Unpublished

manuscript.

Kaufmann, Franz Xaver. 1988. “Christentum und Wohlfahrsstaat.” Zeitschrift fur

Sozialreform. 34(2): 65-89.

Kiernan, Kathleen, Hilary Land, and Jane Lewis. 1998. Lone Motherhood in Twentieth-Century

Britain. Oxford: Clarendon Press.

Knijn, Trudie. 1994. “Fish Without Bikes: Revision of the Dutch Welfare State and Its

Consequences for the (In)Dependence of Single Mothers.” Social Politics 1: 83-105.

_____. 1998. “Social Care in the Netherlands.” Pp. 85-110 in Gender, Social Care, and Welfare

State Restructuring in Europe, ed. J. Lewis. Aldershot: Ashgate.

34 Knijn, Trudie, and Frits van Wel. 2001. “Careful or Lenient? Welfare Reform for Lone Mothers

in the Netherlands.” Journal of European Social Policy 11: 235-52.

Kolberg, Jon Eivind, and Gosta Esping-Andersen. 1992. “Welfare States and Employment

Regimes.” Pp. 3-35 in The Study of Welfare State Regimes, ed. J. E. Kolberg. Armonk,

NY: M. E. Sharpe.

Korpi, Walter, and Joakim Palme. 1998. “The Paradoxes of Redistribution and Strategies of

Equality: Welfare State Institutions, Inequality, and Poverty in the Western Countries.”

American Sociological Review 63: 661-87.

Langan, Mary, and Ilona Ostner. 1991. “Gender and Welfare: Towards a Comparative

Framework..” Pp. 126-150 in Towards a European Welfare State?, ed. G. Room. Bristol,

England: SAUS Publications.

Leira, Arnlaug. 1992. Welfare States and Working Mothers: The Scandinavian Experience. New

York: Cambridge University Press.

Leitner, Sigrid. 2003. “Varieties of Familialism: The Caring Function of the Family in

Comparative Perspective.” European Societies 5: 353-75.

Lessenich, Stephan, and Ilona Ostner. 1995. “Die Institutionelle Dynamik ‘dritter Wege’: Zur

Entwicklung der Familienpolitik in katholischen Wohlfahrtstaaten am Beispiel

Deutschlands und Frankreichs.” Zeitschrift furSozialereform 41: 780-803.

Lewis, Jane, ed. 1997. Lone Mothers in European Welfare Regimes:Shifting Policy Logics.

London: Jessica Kingsley Publishers.

Lewis, Jane, and Ilona Ostner. 1991. “Gender and the Evolution of European Social Policies.”

Center for European Studies Workshop “Emergent Supranational Social Policy: The

EC’s Social Dimension in Comparative Perspective,” Cambridge, MA (November).

35 McLanahan, Sara, and Irwin Garfinkel. 1995. “Single-Mother Families and Social Policy:

Lessons for the United States from Canada, France, and Sweden.” Pp. 367-83 in Poverty,

Inequality, and the Future of Social Policy, K. McFate, R. Lawson, and W. J. Wilson.

NY: Russell Sage.

Millar, Jane. 1996. “Mothers, Workers, and Wives: Comparing Policy Approaches to Supporting

Lone Mothers.” Pp. 97-113 in Good Enough Mothering?Feminist Perspectives on Lone

Mothering, ed. E. Bortolaia Silva. London: Routledge.

Moller, Stephanie, David Bradley, Evelyne Huber, François Nielsen, and John D Stephens. 2003.

“Determinants of Relative Poverty in Advanced Capitalist Democracies.” American

Sociological Review 68: 22-51.

O’Connor, Julia, Ann Orloff, and Sheila Shaver. 1999. States, Markets, and Families: Gender,

Liberalism, and Social Policy in Australia, Canada, Great Britain, and the United States.

Cambridge: Cambridge University Press.

Organization for Economic Co-Operation and Development (OECD). 1995. “Income

Distribution in OECD Countries: Evidence from the Luxembourg Income Study.” Social

Policy Studies No. 18. Paris, France.

Orloff, Ann Shola. 1993. “Gender and the Social Rights of Citizenship.” American Sociological

Review 58: 303-28.

_____. 1996. “Gender in the Welfare State.” Annual Review of Sociology 22: 51-78.

Ostner, Ilona. 1994. “Back to the Fifties: Gender and Welfare in Unified Germany.” Social

Politics 1: 32-59.

36 _____. 1997. “Lone Mothers in Germany Before and After Unification.” Pp. 21-49 in Lone

Mothers in European Welfare Regimes: Shifting Policy Logics, ed. J. Lewis. London:

Jessica Kingsley Publishers.

Pfau-Effinger, Birgit. 1999. “Change of Family Policies in the Socio-Cultural Context of

European Societies.” Comparative Social Research 18: 135-59.

Rainwater, Lee, and Timonthy J. Smeeding. 2003. Poor Kids in a Rich Country: America’s

Children in Comparative Perspective. New York: Russell Sage.

Sainsbury, Diane. 1995. Gender, Equality, and Welfare States. New York: Cambridge.

Saraceno, Chiara. 1997. “Family Change, Family Policies and the Restructuring of Welfare.” Pp.

81-99 in Family, Market and Community: Equity and Efficiency in Social Policy. Paris:

OECD.

Schultheis, Franz. 1992. “Inerties Structurelles et Ambivalences Ideologiques: La Protection

Sociale Allemande Face Aux Nouveaux ‘Risques Familiaux’.” Recherches Sociale 122:

144-57.

Taylor-Gooby, Peter. 1991. “Welfare State Regimes and Welfare Citizenship.” Journal of

European Social Policy 1: 93-105.

van Kersbergen, Kees. 1994. “The Distinctiveness of Christian Democracy.” Pp. 31-50 in

Christian Democracy in Europe: A Comparative Perspective, ed. D. Hanley. London:

Pinter Publishers.

_____. 1995. Social Capitalism: A Study of Christian Democracy and the Welfare State. New

York: Routledge.

37 Van Kersbergen, Kees, and Uwe Becker. 1988. “The Netherlands: A Passive Social Democratic

Welfare State in a Christian Democratic Ruled Society.” Journal of Social Policy 17:

477-99.

38

Table 1. Leitner's Combinations of Familialization and De- Familialization Defamilialization Familialization Strong Weak Strong Optional Familialism Explicit Familialism Weak De-Familialism Implicit Familialism

Table 2. Classifications of Continental European “Conservative” Nations Belgium France Germany Netherlands Researcher Esping- Conservative Conservative Conservative Conservative Andersen

Lewis & Ostner Moderate Moderate Strong Strong Breadwinner Breadwinner Breadwinner Breadwinner

Millar High High High Low Employment for Employment for Employment for Employment for Married & Single Married & Single Mothers; Married & Mothers Single Mothers Low Single Mothers Employment for Married Mothers

Leitner Optional Optional Explicit Explicit Familialism Familialism Familialism Familialism

2 Table 3. Pre-Tax and Transfer Poverty Rates for Married and Lone Women with No Children, Children, and Young Children Married Women Single Women without with with young without with with young Year children children children children children children Conservative Regime Belgium 1997 13.1 (9) 13.9 (7) 14.5 (3a) 34.5 (9) 50.1 (3) - France 1994 10.2 (7b) 12.1 (3) 16.7 (6) 27.0 (5a) 44.7 (2) 64.5 (1) Germany 1994 7.2 (4) 11.6 (2) 14.6 (3b) 26.6 (4) 62.5 (8) - Netherlands 1994 14.7 (10) 12.5 (4a) 9.8 (2) 35.4 (10) 77.9 (9) - Social Democratic Regime Finland 1995 9.3 (6) 13.3 (6) 19.0 (8a) 25.8 (3) 41.4 (1) - Norway 1995 5.5 (1) 7.8 (1) 9.1 (1) 23.6 (2) 59.3 (5) 76.9 (5) Sweden 1995 6.0 (2) 12.7 (4b) 16.2 (5) 28.9 (7) 50.8 (4) 65.5 (2) Liberal Regime Canada 1994 8.8 (5) 16.5 (9) 19.2 (8b) 29.1 (8) 59.6 (6) 71.7 (3) United Kingdom 1995 10.1 (7a) 21.0 (10) 25.7 (10) 27.2 (5b) 79.1 (10) 89.0 (6) United States 1994 6.9 (3) 15.1 (8) 18.1 (7) 23.1 (1) 60.3 (7) 72.6 (4) AVERAGE 9.2 13.6 16.3 28.1 58.6 73.4

Conservative Regime Germany 2000 7.0 (2d) 9.3 (3a) 11.1 (3) 23.3 (5a) 50.6 (3) - Netherlands 1999 9.7 (7) 8.6 (2) 8.2 (1) 21.1 (3) 48.7 (2) - Social Democratic Regime Finland 2000 7.1 (2e) 12.4 (5) 17.0 (5) 22.2 (4) 40.2 (1) - Norway 2000 5.0 (1) 6.8 (1) 8.9 (2) 23.5 (5b) 54.4 (6) 73.8 (3) Sweden 2000 6.5 (2a) 9.4 (3b) 13.3 (4) 24.7 (7) 52.0 (4) 65.6 (2) Liberal Regime United Kingdom 1999 6.8 (2c) 15.9 (7) 19.8 (7) 19.7 (2) 65.8 (7) 77.4 (4) United States 2000 6.7 (2b) 15.6 (6) 19.3 (6) 17.7 (1) 53.5 (5) 58.0 (1)

AVERAGE 7.0 11.1 13.9 21.7 52.2 68.7 Note: Numbers in parentheses represent country rank from lowest poverty rate (1) to the highest Table 4. Post-Tax and Transfer Poverty Rates for Married and Lone Women with No Children, Children, and Young Children Married Women Single Women without with with young without with with young Year children children children children children children Conservative Regime Belgium 1997 3.6(5) 6.4 (6) 9.0 (7) 16.1 (8) 11.6 (4) - France 1994 3.9(6) 5.6 (5) 6.9 (6) 11.7 (6) 22.8 (5) 33.9 (3) Germany 1994 1.8(4) 5.2 (4) 6.4 (4a) 13.7 (7) 43.1 (9) - Netherlands 1994 4.6(8) 6.9 (7) 6.4 (4b) 9.2 (4) 25.0 (6) - Social Democratic Regime Finland 1995 1.3 (1b) 1.5 (1b) 1.6 (1a) 6.7 (2a) 3.2 (1) - Norway 1995 1.2 (1a) 1.7 (1c) 2.0 (3) 6.9 (2b) 10.9 (3) 13.7 (2) Sweden 1995 1.5 (1c) 1.4 (1a) 1.7 (1b) 4.9 (1) 5.7 (2) 9.0 (1) Liberal Regime Canada 1994 4.2(7) 8.7 (8) 10.7 (8) 17.4 (9) 39.7 (8) 52.2 (5) United Kingdom 1995 5.3(10) 11.9 (9a) 15.1(9) 10.6(5) 38.5 (7) 50.0 (4) United States 1994 4.9 (9) 12.1 (9b) 15.4 (10) 18.6 (10) 48.9 (10) 61.1 (6) AVERAGE 3.2 6.2 7.5 11.6 24.9 36.7

Conservative Regime Germany 2000 2.5(4) 2.2 (1d) 4.1 (4) 18.8 (7) 31.6 (6) - Netherlands 1999 6.4 (7) 7.7 (5) 8.7 (5) 9.1 (2b) 30.9 (5) - Social Democratic Regime Finland 2000 1.1 (1) 1.9 (1a) 2.1 (1a) 8.1 (1) 7.7 (1) - Norway 2000 1.4 (2) 1.9 (1b) 2.3 (1b) 9.1 (2a) 10.5 (2a) 19.5 (1) Sweden 2000 2.1 (3) 2.1 (1c) 3.1 (3) 9.4 (4) 10.5 (2b) 21.9 (2) Liberal Regime United Kingdom 1999 4.0(5) 8.7 (6) 11.6 (6) 11.9 (5) 29.2 (4) 38.3 (3) United States 2000 5.3(6) 12.1 (7) 15.1 (7) 18.1 (6) 42.8 (7) 48.7 (4) AVERAGE 3.3 5.2 6.7 12.1 23.3 32.1 Note: Numbers in parentheses represent country rank from lowest poverty rate (1) to the highest

Table 5. Post-Tax-and-Transfer Poverty Rates Lower for Single Mothers Higher for Single Mothers Lower for Married Mothers Finland, France, Norway, Germany Sweden Higher for Married Mothers Belgium Canada, Netherlands, UK, US

Table 6: Employment Rates for Married and Single Mothers, Early 1990s Married/Cohabiting Mothers Single Mothers % Employed % Employed All who are Full- All who are Full- Full-time Part-time employed time Full-time Part-time employed time Conservative Regime Belgium 1992 36 22 61 59 52 16 68 76 Finland 1993 62 8 70 89 61 4 65 94 France 1992 49 20 68 72 67 15 82 82 Germany 1992 21 20 41 51 28 12 40 70 Netherlands 1994 13 39 52 25 16 24 40 40

Social Democratic Regime Norway 1991 40 37 77 52 44 7 51 72 Sweden 1994 42 38 80 53 41 29 70 59

Liberal Regime United Kingdom 1990-92 21 41 62 34 17 24 41 41 United States 1992 45 19 64 70 47 13 60 78

AVERAGE 36.56 27.11 63.89 56.11 41.44 16.00 57.44 68.00 Source: Chiara Saraceno. 1997. “Family Change, Family Policies and the Restructuring of Welfare.” Family, Market, and Community: Equity and Efficiency in Social Policy. OECD.

Table 7. Labor Participation Rates for Mothers

High for Single Mothers Low for Single Mothers High for Married Mothers Finland, France, Sweden, Norway US Low for Married Mothers Belgium Germany, Netherlands, UK

Table 8. The Strength of Family Policies Strength of Policy Parental Leave Childcare Policy Childcare Policy for Children <3 for Children 3-6

Weak Canada Canada Canada UK Germany Germany US Netherlands Netherlands UK UK US US

Strong Belgium Belgium Belgium Finland Finland Finland France France France Germany Norway Norway Netherlands Sweden Sweden Norway Sweden

Source: Janet C. Gornick, Marcia K. Meyers, and Katherin E. Ross. 1997. “Supporting the Employment of Mothers: Policy Variation Across Fourteen Welfare States.” Journal of European Social Policy. 7(1): 45- 70 and Janet Gornick and Marcia K. Meyers. 2003. Families that Work: Policies for Reconciling Parenthood and Employment. New York: Russell Sage.

2

Table 9. Leitner's Model of De-Familialization and Familialism Defamilialization Familialization Strong Weak Strong Belgium, Finland, France, Sweden, Netherlands, Norway Germany Weak Canada, US, UK

3 Figure One: Poverty Rates by Policy Approach

60

50

40

Optional Familialism 30 Explicit Familialism Implicit Familialism

20

10

0 with children with young children with children with young children

4

Appendix A: Numbers of Cases for Categories by Nation Married/ Married/ Married/ Single Cohabiting Cohabiting Cohabiting Single Single Women Women Women Women Women Women With Without With With Young Without With Young Country Year Children Children Children Children Children Children Belgium 1997 (1061) (1261) (541) (35) (111) (19) Canada 1994 (6864) (10771) (4859) (2744) (1958)(702) Finland 1995 (1908) (3297) (1406) (539) (287) (50) France 1994 (1885) (3506) (1641) (754) (393)(107) Germany 1994 (1333) (1899) (738) (363) (186) (45) Netherland 1994 (1171) (1569) (715) (356) (130) (27) Norway 1995 (1964) (2979) (1261) (454) (267) (82) Sweden 1995 (2492) (3899) (2001) (885) (471) (194) United Kindom 1995 (1202) (1747) (863) (434) (398) (144) United States 1994 (8540) (14817) (7087) (4489) (4081) (1453) AVERAGE (2842) (4575) (2111) (1105) (828) (282)

Finland 2000 (2396) (3411) (1384) (544) (255) (46) Germany 2000 (1228) (1660) (678) (384) (180) (41) Netherland 1999 (1132) (1437) (663) (294) (104) (13) Norway 2000 (2857) (4282) (1862) (712) (428) (102) Sweden 2000 (2371) (3349) (1388) (846) (400) (89) United Kingdom 1999 (4173) (5376) (2497) (1351) (1000) (292) United States 2000 (7915) (12673) (5838) (3910) (2706) (886) AVERAGE (3717) (5731) (2612) (1515) (725) (210)

5